The radial-symmetric hydra and the evolution of the bilateral body ...

This is a preprint of an article published in BioEssays 24, 185-191 (2002). URL: http: //www3.interscience.wiley.com//journal/34201/home

The radial-symmetric hydra and the evolution of the bilateral body plan: an old body became a young brain1

Hans Meinhardt, Max-Planck-Institut f¨ ur Entwicklungsbiologie, Spemannstr. 35, D- 72076T¨ ubingen, Germany; e-mail: [email protected]

The radial symmetric cnidarians are regarded as being close to the common metazoan ancestor before bilaterality were invented. It is proposed that a large fraction of this gastrula-like organism gave rise to the head of more evolved organisms. The trunk were added later in evolution from an unfolding of a narrow zone between the tentacles and the blastoporus. This implies that, counter intuitively, the foot of the hydra corresponds to the most anterior part (forebrain and heart) while the opening of the gastric column gave rise to the anus. Two fundamentally different modes of midline formation evolved. In vertebrates, the organizer attracts cells from the both sides of the marginal zone. These leave the organizer as a unified band. The midline is formed sequentially from anterior to posterior. In insects, the midline forms opposite of a dorsal repelling center, i.e., on the ventral side. This occurs more or less simultaneously.

of the spatial organization of vertebrates and arthropods. Using the term ‘hydra-like’ for the common ancestor is not to overlook the long evolutionary history of the current hydra with its highly elaborated features such as the nematocyst cell. The term refers only to the elementary body plan. It is assumed that the expression patterns of gene families employed for the basal organization of body axes have been preserved. In contrast, the morphological features these genes control in contemporary animals may have changed completely due to the large phylogenetic and temporal distance. There is an increasing evidence that the oral opening of a hydra corresponds to the blastopore of a vertebrate gastrula. The brachyury homologue of hydra, Hy-bra1 is expressed around the oral opening1 (Fig. 1), corresponding to the brachyury expression in the in the marginal zone in Xenopus 2 next to the endoderm. A basic question is how the dimension extending from the oral opening to the opposite region of the radial symmetric ancestor, conventionally called the head-foot axis of hydra, is related to the body axes of the bilateral animals.

A central issue of evolutionary biology is how the two major body axes of higher organisms, anteroposterior and dorsoventral, evolved from the single oral-aboral axis of simpler radial-symmetric organisms several hundred millions years ago. In recent years, evidence accumulated that the basic regulatory pathways are highly conserved. In the patterning of the freshwater polyp hydra, molecules have been found that participate in higher organisms in the generation of both axes. For this reason, hydra is most likely closely related to the putative common radial-symmetric ancestor of coelenterates and higher animals. Comparative analysis with respect of axis formation and the molecules involved are expected to shed light on the evolution 1 For

Two conflicting hypothesis have been put forward (see ref. 3-5 for review). According to the protostomia - deuterostomia concept, a widening of the blastoporus took place in one dimension, resulting in a slit-like geometry. By this ‘Amphistomy’5 , the blastoporus obtained a long extension that forms the future midline. One end of this slit becomes anterior, the other posterior. Either the mouth (protostomia) or the anus (deuterostomia) remained as an opening of the original gastric cavity. The paired arrangement of the nerve cord on both sides of the midline is assumed to be derived from the ring of high nerve cell density around the blastoporus4−6 . According to this

copyright reasons this article can be provided here only in the pre-publication form

1

regions of these two genes become separated from each other during the sequential anterior to posterior trunk formation. Their initial border can be regarded as the head/trunk border. In hydra, these two genes are also expressed adjacent to each other in a narrow zone. It is located between the tentacles and the hypostome; the goosecoid homologue Cngsc closer to the tentacles, HyBra1 closer to the opening8,1 . This suggests that most of the AP body axis of contemporary higher organisms, the trunk, is derived from a narrow ring of the putative hydra-like ancestor located between the tentacle zone and the opening of the gastric column (Fig. 1). In contrast, the major fraction of the original gastrula-like organism became converted into the head and especially into the brain of higher animals. With this orientation of the axes, the foot of the hydra corresponds to the most anterior part of the brain while a region between tentacle zone and the cone-shaped hypostome would correspond to the posterior end of the brain. This scheme accounts for the fact that the AP axis of a higher organism consist of two parts with very different properties. One part, the head, would be derived from most of the original gastrea-like organisms, whereas the second part, the trunk and the posterior end, resulted from an unfolding of a very small portion of the original radial-symmetric organism. Only during the evolution of the latter process, the notochord and the Hox gene code was invented. It is, therefore, not surprising that in hydra the typical pattern of Hox gene expression has not been found although HOX genes exists14−17 : the trunk was not yet there. The sequential elongation of the trunk from a narrow ring during evolution and the elaboration of the sequential co-linear Hox gene pattern arose presumably in a coupled way. In many higher organisms such a sequential elongation of the trunk is recapitulated during the ontogenetic development. Assuming that the body column of hydra corresponds to the fore- and midbrain of vertebrates, it is expected that corresponding gene expressions can be found there. This is indeed the case. The Hox gene Cnox-2 is expressed in the body column of the hydra, but not in the hypostome nor in the tentacle region17 . Homologues genes are expressed in the brain of the mouse18 and in the cerebral vesicle of amphioxus19 , a structure homologous to the vertebrate fore- and midbrain. The gene Otx is expressed in the hydra body column including the tentacle zone8 . Its homologue in vertebrates, expressed in the front part of the brain, is crucial for the formation of midbrain/hindbrain organizing region (ref. 20-23 for review). In vertebrates, the region posterior to Otx expression forms the hindbrain and has already a notochord with brachyury expression. Thus, the expression border of Otx at the oral side, abutting the region of HyGsc expression, marks the posterior end the of old part of the body that evolved into the fore- and midbrain. In Hydractinia, a colonial marine relative of hydra, an ems- related gene has been found. It is expressed between the tentacle zone and the oral opening24 . According

concept, the mouth opening of the hydra-like ancestor became after elongation the future midline and thus the neural side of higher organisms (i.e., dorsal in vertebrates and ventral in arthropods and insects). To accomplish this ring-slit transition, a signal for “most anterior” and “most posterior” must be generated on the ring at opposing positions. This hypothesis implies that the neural – anti-neural (mediolateral) axis was first in evolution and the anteroposterior axis was the new invention. An alternative view is that the oral – aboral axis of a radial-symmetric hydra-like ancestors gave rise to the anteroposterior axis of higher animals. The elongated column of the hydra, the food intake at the oral ‘head’ and fixation to the ground with the aboral peduncle (‘foot’) makes this assignment reasonable and the use of the terms ‘head’ and ‘foot’ are suggestive. Many genes expressed in the head of higher organisms are expressed in the hydra head. Examples are Otx 7 , Aristaless 8 and goosecoid 9 . This suggest that the expression patterns of genes characteristic for head development became established before the invention of bilaterality (for review ref. 10). In this view, the AP axis of higher animals would be primary and mediolateral the newly invented one. Such an assignment does not necessary imply that the orientation of the AP axis of higher animals corresponds to the head-to-foot axis of hydra. In many organisms the blastopore becomes the anus11,12 . It has been pointed out by Wu and Lengyel13 , that genes such as caudal, wnt and brachyury are expressed at the posterior pole of Drosophila and, likewise, around the blastopore of vertebrates. In hydra, wnt and brachyury are also expressed around the opening of the gastric column, suggesting that this part corresponds to blastoporus and thus to the most posterior structure of higher organisms. How can this be reconciled with the evidence that head-specific genes of bilateral animals are expressed in the head of hydra? The explanation proposed is that the most of the body of the hydra-like ancestor gave rise fore- and midbrain in vertebrates. This region ranges in hydra from the foot very far into the body and includes the tentacle area (which is usually considered as part of the head region). Only the narrow zone between tentacles and the gastric opening gave rise to the trunk and anus. Therefore, gene systems that were used in the hydra-like ancestor to organize most of the whole body can be found in contemporary animals to organize the brain. The proposition is supported by the following comparison of genes expressed along the AP axes in higher animals and in hydra respectively.

The ancient gastrula-like organism develop into the head; the trunk is added in the course of time In vertebrates, goosecoid and brachyury are initially expressed in adjacent regions. While goosecoid is involved in head formation, the pan-mesodermal gene brachyury is crucial for trunk and tail formation. Thus, the expression

2

H y d ra

(A )

H y d a - lik e a n c e s to r

(B )

P r o s p e c tiv e r e g io n s

B r a in

(F )

T ru n k

V e rte b ra te s

In s e c ts

(C )

(G )

(D )

(H ) (I)

(E )

A

F ig . 1 : P r o p o s e d e q u iv a le n c e p a tte r n s a lo n g th e A P a x is o f a h y d r a - lik e a n c e s to r a n d c o n te m p o r a r y b ila te r ia n s : th e fo o t c o r r e s p o n d s to th e m o s t a n te r io r s tr u c tu r e s in c lu d in g th e fo r e b r a in a n d h e a r t ( N k 2 r e la te d g e n e s , v io le t) ; th e c e n tr a l p a r t o f th e g a s tr ic c o lu m n ( O tx , b lu e ) to c e n tr a l p a r t o f th e b r a in ; th e b o r d e r o f te n ta c le r e g io n ( A r is ta le s s , d e e p b lu e ) a n d th e h y p o s to m e ( G s c , y e llo w ) r e g io n to th e m id /h in d b r a in r e g io n . O n ly th e h y p o s to m e w ith b r a c h y u r y a n d W n t (re d , g re e n ) a p p e a rs a s “a n c e s to r a re a ” o f th e tr u n k , th e o p e n in g b e c a m e th e a n u s . In v e r te b r a te s ( C - E ) , th e p r im a r y o r g a n iz in g r e g io n ( ß - c a te n in , T c f; g r e e n ) a n d G s c b e c a m e c o n tr a c te d to a p a tc h - lik e e x te n s io n ( a r r o w ) n e a r th e b la s to p o r u s fo r m in g th e ( S p e m a n n - lik e ) o r g a n iz e r w ith a n a n te r io r ( y e llo w ) a n d p o s te r io r p a r t ( g r e e n ) . G s c e x p r e s s in g c e lls ( y e llo w ) m o v e a n te r io r to fo r m th e p r e c h o r d a l p la te , s p e c ify in g d o r s a l in th e b r a in r e g io n . C e lls o f th e m a r g in a l z o n e in c lu d in g b r a c h y u r y - e x p r e s s in g c e lls ( r e d ) m o v e fr o m b o th s id e s to w a r d s th e o r g a n iz e r a n d le a v e th e m a s a u n if ie d b a n d . M id lin e a n d t r u n k a r e f o r m e d in a s e q u e n tia l w a y fr o m a n te r io r ( D ) to p o s te r io r ( E ) . T h u s , b r a c h y u r y e x p r e s s in g c e lls fo r m fir s t a r in g a r o u n d t h e b la s t o p o r u s ( C ) a n d p a r t ic ip a t e la te r in n o t o c h o r d fo r m a tio n . In in s e c ts ( F - I) , in itia l m id lin e fo r m a tio n ( o r a n g e ) is g o v e r n e d b y a r e p e llin g in flu e n c e fr o m th e d o r s a l s id e ( v io le t, s e e in s e r t fo r D r o s o p h ila ,) . T h is c a n p r o c e e d s im u lta n e o u s ly a lo n g th e w h o le v e n tr a l e x te n s io n . B r a c h y u r y / b r a c h y e n te r o n r e m a in s r e s tr ic te d to th e p o s te r io r p o le . D u r in g fu r th e r p o s te r io r o u tg r o w th , e .g ., in s h o r t g e r m in s e c ts e s ta b lis h e d D V o r g a n iz a tio n is m a in ta in e d a n d th e m id lin e b e c o m e s e lo n g a te d . W ith in c r e a s in g d is ta n c e o f t h e c e lls t o t h e z o n e o f w in g le s s ( g r e e n ) e x p r e s s io n a t p o s te r io r p o le , n e w s tr ip e s o f w in g le s s e x p r e s s io n c a n b e in itia te d . T h e r e s u ltin g r e p e titiv e p a tte r n p a r tic ip a te s in s e g m e n t fo r m a tio n ( H - I) .

P

D V A

P

to the proposed assignment of the A-P axis, this is posterior to Otx. This fits with the AP-arrangement of both genes in Drosophila where ems is also expressed posterior to Otx/Odt25 . Several genes are known that play a role in foot formation of hydra26,27 . An NK2 related gene is expressed in a graded way towards the foot26 . In zebrafish, an Nk2 gene demarcates a zone of neuronal differentiation in the embryonic forebrain28 . Corresponding expressions at very anterior (-ventral) positions has been observed in Amphioxus 29 , Xenopus 30,31 , mouse32 and chick33 . In Drosophila, an NK2 related gene is initially expressed only in future brain region of the embryo34 . In C.elegans, an Nk2 gene is involved in pharynx formation35 . These observations are compatible with the view that the region corresponding to the foot of hydra gave rise to the anterior-most portion of the body. Nkx2.5 related genes are involved in heart formation, also a very anterior (-ventral) structure36,37 . Recently it has been shown for hydra that ß-catenin, Tcf and Wnt are expressed in the hypostome, with Wnt restricted to the tip38 . These molecules are known since long to be involved in the vertebrate organizer39 . A WNT gene , WNT-3a, is required for posterior development in the mouse40 . Without WNT-3a, no node and no

streak is formed. Likewise in Amphioxus, Wnt is expressed at the blastoporus and this expression remains there as this regions develops into the most posterior part of the organism2 . In Drosophila as well as in Tribolium, WNT is expressed around the posterior pole41 . Wu and Lengyel13 emphasized the similarities between posterior wnt/brachyury expression in Drosophila and around the blastoporus of vertebrates. The corresponding expression patterns in hydra are in full agreement with the stipulation that the mouth opening became the most posterior part, the anus.

In Cnidarians that go through a larval stage, the posterior part of the larva gives rise to the ‘head’ of the final polyp42 . This has been interpreted as a polarity reversal. According to the proposed orientation of AP axis no such polarity reversal occurs: the posterior part of a larva gives rise to the posterior part of the polyp, the gastric opening. Thus, not only the adult animal but also the larva can provide information about the ancestral gastrula.

3

Bilateral symmetry and the formation of a midline

ter crossing the midline46 ): they are no longer attracted. Therefore, cells from both sides of the ring move towards the organizer and leave this region as a unified band – the future midline (Fig. 1). It has a 90◦ orientation in relation to the original ring (for modelling, see ref 47). This is a special mode of the movement generally known as conversion-extension movement48 . The passage of cells through the organizer has been most clearly demonstrated for the chick49 . Cells from the organizer may be added to the band and contribute to structures such as the notochord and floor plate. In the course of time this leads to a conversion of the ring- into a rod-shaped arrangement of the mesoderm, most clearly seen in the brachyury expression during gastrulation2,50 . Thus, the organizer (or node) can be regarded as a spot-like device that elongates the midline by first attracting and then reprogramming (and perhaps eventually repelling) the migrating cells. The organizer is therefore always located at the tip of the midline, causing its sequential elongation from anterior to posterior. The time at which cells leave the organizer fixes presumably their anteroposterior identity by stopping a sequential activation of more posterior HOX-genes. This midline formation starts with notochord formation at the contemporary hindbrain level posterior to the Otx expression. Therefore, a separate mechanism is required that allows a mediolateral organization of the foreand midbrain. This is proposed to be achieved by the anterior movement cells to form the prechordal plate. Goosecoid -expressiong cells, originally localized at the organizer region move anteriorly and are able to accomplish an anterior/dorsal re-specification51 . Thus, goosecoid may be a part of the expected machinery required for the anterior elongation of the midline into the fore- and midbrain region. During trunk formation, the distance of the posterior organizer to anterior regions increases to and becomes thus less and less precise in supplying positional information for the organization of the brain. This is presumably the reason why the fading posterior organizer becomes replaced by a new one that remains in place - the istmic organizer at the midbrain/hindbrain border20−23 .

The transition from a radial-symmetric cup-shaped gastrula to a bilateral-symmetric organism requires a symmetry breaking as indicated by the formation of a Spemann-type organizing region close the blastoporal margin. However, a local high concentration on a ring is on itself insufficient to generate two axes that have an orthogonal orientation relative to each other: from a single high point the concentration decreases in all directions; therefore, the early amphibian organizer marks both a most dorsal and a most anterior point of the trunk. At this stage, these axes are not yet separated. The crucial step towards a bilateral-symmetric coordinate system for animals with an extended trunk is the generation of a midline such that positional information for the mediolateral extension can be generated. This requires the formation of a stripe-like organizing region as given by the notochord of vertebrates or by the ventral midline of insects. A single straight stripe cannot be generated by a single pattern forming mechanism. The reason is easy to understand. A single spot-like local high concentration can be generated by local self-enhancement and a strong long-range inhibition43 . Stripe formation occurs if the selfenhancement has an upper bound otherwise the stripe will decay into separate patches44,45 . When initiated by uniform initial conditions, the resulting patterns consist of somewhat meandering and branching stripes such as seen in the pigmentation patterns on some tropical fishes or on the skin of zebras. Due to the restricted inhibition, the width of the stripes is of the same order as the width of the interstripe regions. A single stripe would require a strong lateral inhibition, but such an inhibition would cause a decay of the stripe into individual patches. However, by an organizing influence of a single spot-like organizer it can be achieved that only a single stripe is formed and that this stripe is straight. It is proposed that this patterning problem has been solved in two different ways: in vertebrates the organizer causes a local initiation and elongation of the midline. In contrast, insects a repulsion of a dorsal organizer restrict the midline to the ventral side of the embryo. Thus, an understanding of this process may provide a key for this fundamental branching during evolution.

Midline formation in insects: a repelling influence from dorsal leads to ventral midline formation In Drosophila a spot-like organizing region is first formed dorsally, i.e., opposite to the future ventral midline (Fig. 1F). It is based on a signalling between the oocyte nucleus and the follicle cells by members of the EGF pathway – a signalling system that leads also to the dorsal appendages52−54 . This inhibition restricts pipe expression to the ventral follicle cells55,56 . The asymmetry is sufficient to generate a single ventral midline. In Drosophila this mode of midline formation does not occur in a sequential fashion as in vertebrates but can happen more or less simultaneously in the early embryo over the whole AP ex-

Midline formation in vertebrates: the spot-like organizer elongates the midline In vertebrates, midline formation can be explained by the assumption that the organizer, located on the germ ring (in zebra fish) or on the marginal zone (Xenopus) attracts cells of these rings. Cells move toward the organizer and, after being close once, they change their behaviour (similar as some nerve cells change their orientation af-

4

the early established mediolateral pattern is posteriorely elongated. This is in contrast to vertebrates, where cells obtain the dorsal reference character only as they pass the node or organizer.

tension. A similar patterning mechanism is also involved in the short germ insect Tribolium 57 , although there an elongation of the midline has to take place at later stages. So far, these two holometabolous (that is, highly derived) insects are the only systems for which detailed information for DV patterning are available. It is assumed, however, that this mode of DV patterning – localization of the midline to the opposite side by a repulsion emanated from a spot-like region – is representative for patterning primitive arthropods (see also below). This scheme provides possibly a new inroad to the apparent inversion of the DV axis in insects and vertebrates58−60 since in insects midline formation occurs opposite to the spot-like organizing region while in vertebrates, it is formed on the same site as the organizer. Not yet clear is of how these different modes of midline formation led to the employment of the same genes for the actual dorsoventral pattern, dpp/short gastrulation in insects and chordin/BMP in vertebrates. This model of midline formation explains why in insects an anterior movement of cells like that for prechordal plate formation in vertebrates does not occur. In insects, the midline starts at the most anterior level, and not, as in vertebrates, only posterior to Otx. Therefore, an anterior elongation of the midline is not required for the DV organization of the insect brain. This model accounts for the different expression patterns of Brachyury/brachyenteron in vertebrates and insects (Fig. 1): In vertebrates it is expressed in the notochord, i.e., at the most dorsal level and along the whole AP axis of the trunk. In contrast, in insects, it is expressed at a very posterior position in a ring, i.e., localized at a specific AP level but occupies but all DV levels61,13 . For this reason the function of brachyury as tail gene has been questioned62 . According to the model proposed, in vertebrates, brachyury participates in the ring – rod conversion and becomes, therefore, expressed along the whole AP axis. In contrast, in insects the expression of the homologous brachyenteron remained as it was in the putative hydra-like ancestor: a ring of activity close to the most posterior pole of the animal. While in vertebrates, brachyury is essential for mesoderm formation, it is never expressed in the mesoderm of insects, underlining the stipulation that the different modes of midline formation were a fundamental branch during evolution.

The generation of the primary AP axis during hydra embryogenesis In hydra, after some initial cell divisions, first a hollow sphere is formed that consists of a single-layered cell sheet. Cells delaminate from this sheet, move to the inside and form eventually the endoderm65 . Although the molecular basis of this process is not yet known, it is expected to be based on a conserved pathway in which known genes responsible for endoderm formation play a role, for instance, VegT 66,67 . On theoretical grounds it is expected that autoregulatory loops are involved in this activation (e.g., VegT and derriere 68 or other nodal -related factors). Minor changes in the parameters can either lead to a saltand-pepper – like pattern as in hydra or to large coherent patches as required for determination of hemispheres as in amphibians47 . After the formation of the concentric diploblastic ectodermal/endodermal spheres a genuine de-novo patterning process is required to generate an AP axis. The capability for that is demonstrated by experiments in which hydra body columns are disaggregated into individual cells. After their re-aggregation first a sorting out of ecto- and endodermal cells occurs, recapitulating early stages of embryonic development. Eventually viable polyps are formed69 . In the spherical aggregates, high levels of ß-catenin and TCF expression are turned on35 . Initially this expression is uniform as expected on theoretical consideration70,43 . Later discrete maxima are formed. This suggest, that ß-catenin/TCF is part of a pattern forming reaction that defines the posterior pole. At the highest points of this pattern, more localized spots of Wnt expression emerges, causing presumably the actual gastric openings of this bi-layered hollow sphere. In hydra, in addition to the primary signal for the most posterior pole and blastoporus, a further pattern forming system becomes active at the antipodal site causing foot formation. Analogously in vertebrates, in addition to the organizer, a separate pattern forming system seems to be required to form the anterior part of the head (for review see ref. 71-73). Once these two counter-oriented gradients exist, a reliable subdivision into discrete zones is possible. A corresponding model for hydra70 found recently support from observations at the molecular level8,26 . On the basis of the proposed homology of foot formation in hydra and of the anterior structures in head development common pathways are expected. The model allows several predictions. Caudal 13 is expressed both in vertebrates and insects. It was therefore also likely present in the common precursor, suggesting that it can be also found in Hydra. It is expected at the most posterior position, the hypostome. In vertebrates,

A-P Elongation of the DV axis in annelids and arthropods In many segmented animals, the zone around posterior pole becomes a growth zone (Fig. 1). However, this elongation is different from that in vertebrates. This is most easily understandable by regarding segment formation in leech63 and some lower crustaceans64 . The trunk pattern is generated from bundles of blast cells that bud off small cells which give rise to the germ band. The blast cell bundles have a fixed mediolateral pattern as do the small cells they give rise to. Thus, during the posterior outgrowth

5

8. Smith, K.M., Gee, L. and Bode, H.R. HyAlx, an aristaless-related gene, is involved in tentacle formation in hydra. Development 2000; 127: 4743-4752. 9. Broun, M., Sokol, S. and Bode, H.R. Cngsc, a homologue of goosecoid, participates in the patterning of the head, and is expressed in the organizer region of hydra. Development 1999,126: 5245-5254. 10. Galliot, B. and Miller, D. (2000). Origin of anterior patterning - how old is our head?. Trends Genet 2000: 16: 1-5. 11. Vogt, W. Gestaltungsanalyse am Amphibienkeim mit ortlicher Vitalf¨ arbung. Roux Arch. Entwicklungsmech. ¨ Org. 1929; 120: 384-706. 12. Holland, L.Z., Holland, N.D. and Schubert, M. Developmental expression of amphiwnt1, an amphioxus gene in the wnt1/wingless subfamily. Dev. Genes & Evol. 2000; 210: 522-524. 13. Wu, L.H. and Lengyel, J.A. Role of caudal in hindgut specification and gastrulation suggests homology between Drosophila amnioproctodeal invagination and vertebrate blastopore. Development 1998; 125: 24332442. 14. Gauchat, D., Mazet, F., Berney, C., Schummer, M., Kreger, S., Pawlowski, J. and Galliot, B. (2000). Evolution of Antp-class genes and differential expression of hydra hox/parahox genes in anterior patterning. PNAS 2000; 97: 4493-4498. 15. Kuhn, K., Streit, B. and Schierwater, B. Isolation of hox genes from the scyphozoan Cassiopeia xamachana: Implications for the early evolution of hox genes. J Exp. Zool 1999; 285: 63-75. 16. Martinez, D.E., Bridge, D., Masudanakagawa, L.M. and Cartwright, P. Cnidarian homeoboxes and the zootype. Nature 1998; 393: 748-749. 17. Shenk, M.A., Bode, H.R. and Steele, R.E. Expression of cnox-2, a hom hox homeobox gene in hydra, is correlated with axial pattern-formation. Development 1993; 117: 657-667. 18. Hsieh-Li, H.M., Witte, D.P., Szucsik, J.C., Weinstein, M., Li, H. and Potter, S.S. (1995). Gsh-2, a murine homeobox gene expressed in the developing brain. Mech. Dev. 1995; 50: 177-186. 19. Brooke, N.M., Garcia-Fernandez, J. and Holland, P.W.H. The ParaHox gene cluster is an evolutionary sister of the Hox gene cluster. Nature 1998; 392,920922. 20. Joyner, A.L., Engrailed, wnt and pax genes regulate midbrain hindbrain development. Trends Genetics 1996; 12: 15-20 21. Joyner, A.L., Liu, A. and Millet, S. Otx2, Gbx2 and Fgf8 interact to position and maintain a mid-hindbrain organizer. Curr. Op. Cell Biol. 2000; 12: 736-741. 22. Rhinn, M. and Brand, M. The midbrain-hindbrain boundary organizer. Curr. Op. Neurobiol. 2001; 11: 3442. 23. Wurst, W. and Bally-Cuif, L. (2001). Neural plate patterning: Upstream and downstream of the istmic

several genes have been isolated that are required to induce anterior head structures. The genes Hex, Cerberus and Dickkopf belong to this group (see ref. 71-73). It is expected that homologous genes will be found in Hydra that are required for foot formation. In part, these genes are Wntantagonists. From a model of hydra patterning70 , hypostome and foot formation are expected to be antagonistic processes. Since Wnt/ß-catenin is involved in hypostome formation, the involvement of Wnt antagonists for foot formation is in accordance with this expectation.

Conclusion The freshwater polyp hydra can be regarded as a living fossil that allows insights into axis formation of a putative metazoan ancestor before bilaterality and trunk formation were invented. In relation to contemporary animals, originally a very small anterior part were present that that evolved towards the head. . In contrast, only a small fraction of these hydra-like ancestors gave rise to the trunk that now forms such a large part of the body of higher animals. Cnidarians are among the lowest phyla in which specialized nerve cells exist. In hydra discrete neuronal centers such as ganglia are not yet present. The existing regionalization of the ancestral body column, however, was able to regionalize the brain that evolved from most of this column Acknowledgements I grateful acknowledge many discussion with Christian Beetz and a careful reading of the manuscript by Alfred Gierer

References 1. Technau, U. and Bode, H.R. Hybra1, a brachyury homologue, acts during head formation in hydra. Development 1999; 126: 999-1010. 2. Smith, J.C., Cunliffe, M.A., O’Reilly, J., SchulteMerker, S. and Umbauer, M. Xenopus ’Brachyury’. Seminars Dev. Biol. 1995; 6: 405-410. 3. Holland, L.Z. (2000). Body-plan evolution in the bilateria: early antero-posterior patterning and the deuterostome-protostome dichotomy. Cur. Op. Genet. Dev. 2000; 10: 434-442. 4. Bruce, A.E.E. and Shankland, M. Expression of the head gene Lox22-Otx in the leech Helobdella and the origin of the bilaterian body plan. Dev Biol. 1998; 201: 101-112. 5. Arendt, D. and N¨ ubler-Jung, K. Dorsal or ventral similarities in fate maps and gastrulation patterns in annelids, arthropods and chordates. Mech. Dev. 1997; 61: 7-21. 6. Arendt, D. and N¨ ubler-Jung, K. Rearranging gastrulation in the name of yolk: evolution of gastrulation in yolk-rich amniote eggs. Mech. Dev. 1999; 81: 3-22. 7. Smith, K.M., Gee, L., Blitz, I.L. and Bode, H.R. CnOtx, a member of the Otx gene family, has a role in cell movement in hydra. Dev. Biol. 1999; 212: 392-404.

6

37. Schultheiss, T.M., Xydas, S. and Lassar, A.B. Induction of avian myogenesis by anterior ectoderm. Development 1995,121: 4203-4214 38. Hobmayer, B., Rentzsch, F., Kuhn, K., Happel, C.M., Cramer von Laue, C., Snyder, P., Rothbacher, U. and Holstein, T.W.. Wnt signalling molecules act in axis formation in the diploblastic metazoan hydra. Nature 2000; 407: 186-189. 39. Harland, R. and Gerhart, J. Formation and function of Spemanns organizer. Annual Review Cell and Dev. Biol. 1997; 13: 611-667. 40. Greco, T.L., Takada, S., Newhouse, M.M., McMahon, T.A., McMahon, A.P. and Camper, S.A. Analysis of the vestigial tail mutation demonstrates that wnt-3a gene dosage regulates mouse axial development. Genes Dev. 1996; 10: 313-324. 41. Nagy, L.M. and Carroll, S. (1994). Conservation of wingless patterning functions in the short germ embryo of Tribolium castaneum. Nature 367,460-463 42. Leitz,T. Induction and settlement of metamorphosis of Cnidarian larvae: signals and signal transduction. Invert. Reprod. Dev. 1997; 31: 1-3 43. Meinhardt, H. and Gierer, A. Pattern formation by local self-activation and lateral inhibition. Bioessays 2000; 22: 753-760. 44. Meinhardt, H. Models for positional signalling with application to the dorsoventral patterning of insects and segregation into different cell types. Development (Supplement) 1989; 169-180. 45. Meinhardt, H. Growth and patterning - dynamics of stripe formation. Nature 1995; 376, 722-723. 46. Yuan, W.L., Zhou, L.J., Chen, T.H., Wu, J.Y., Rao, Y. and Ornitz, D.M. The mouse SLIT family: Secreted ligands for ROBO expressed in patterns that suggest a role in morphogenesis and axon guidance. Dev. Biol. 1999: 212: 290-306. 47. Meinhardt, H. Organizer and axes formation as a selforganizing process. Int. J. Dev. Biol. 2001; 45: 177-188. 48. Keller, R., Shih, J. and Sater, A. The cellular basis of the convergence and extension of the Xenopus nueral plate. Dev. Dyn. 1992; 193: 199-217. 49. Joubin, K. and Stern, C.D. Molecular interactions continuously define the organizer during the cell movements of gastrulation. Cell 1999; 98, 559-571. 50. Schulte-Merker, S., Ho, R.K., Herrmann, B.G. and N¨ usslein-Volhard, C. The protein product of the zebrafish homologue and of the mouse T gene is expressed in nuclei of the germ ring and the notochord of the early embryo. Development 1992; 116: 1021-1032. 51. Niehrs, C., Steinbeisser, H. and De Robertis, E.M. Mesodermal patterning by a gradient of the vertebrate homeobox gene goosecoid. Science 1994; 263, 817-820. 52. Roth, S., Neuman-Silberberg, F.S., Barcelo, G. and Sch¨ upbach, T. Cornichon and the egf receptor signaling process are necessary for both anterior-posterior and dorsal-ventral pattern-formation in Drosophila. Cell 1995; 81,967-978.

organizer. Nature Rev. Neurosci. 2001; 2: 99-108. 24. Mokady, O.,Dick, M.H., Lackschewitz, D., Schierwater, B., Buss, L. Over one-half billion years of head conservation? Expression of ems class genes in Hydractinia symbiolongicarpus. PNAS 1998; 95: 3673-3678 25. Finkelstein, R. and Perrimon, N. The moleculargenetics of head development in Drosophilamelanogaster Development 1991; 112: 899-912. 26. Grens, A., Gee, L., Fisher, D.A. and Bode, H.R. Cnnk2, an nk-2 homeobox gene, has a role in patterning the basal end of the axis in hydra. Dev. Biol. 1996; 180: 473-488. 27. Bridge, D.M., Stover, N.A. and Steele, R.E. Expression of a novel receptor tyrosine kinase gene and a paired-like homeobox gene provides evidence of differences in patterning at the oral and aboral ends of hydra. Dev. Biol. 2000; 22: 253-262. 28. Barth, K.A. and Wilson, S.W. Expression of zebrafish nk2.2 is influenced by sonic hedgehog vertebrate hedgehog-1 and demarcates a zone of neuronal differentiation in the embryonic forebrain. Development 1995; 121: 1755-1768. 29. Venkatesh, T.V., Holland, N.D., Holland, L.Z., Su, M.T. and Bodmer, R. Sequence and developmental expression of amphioxus AmphiNk2-1: insights into the evolutionary origin of the vertebrate thyroid gland and forebrain. Dev. Genes Evol. 1999; 209: 254-259. 30. Saha, M.S., Michel, R.B., Gulding, K.M. and Grainger, R.M. A Xenopus homeobox gene defines dorsalventral domains in the developing brain. Development 1993: 118, 193-202. 31. Hollemann, T. and Pieler, T. Xnkx-2.1 : a homeobox gene expressed during early forebrain, lung and thyroid development in Xenopus laevis. Dev. Genes Evol. 2000; 210: 579-581. 32. Shimamura, K. , Hartigan, D. J., Martinez, S. ,Puelles, L. ,Rubenstein, J. L.R. Longitudinal organization of the anterior neural plate and neural tube. Development 1995; 121: 3923-3933 33. Pera, E.M. and Kessel, M. Demarcation of ventral territories by the homeobox gene nkx2.1 during early chick development. Dev. Genes Evol. 1998; 208: 168171. 34. Zaffran, S., Das, G. and Frasch, M.The nk-2 homeobox gene scarecrow (scro) is expressed in pharynx, ventral nerve cord and brain of Drosophila embryos. Mech. Dev. 2000; 94: 237-241. 35. Okkema, P.G. and Fire, A. The caenorhabditis-elegans nk-2 class homeoprotein ceh-22 is involved in combinatorial activation of gene-expression in pharyngeal muscle. Development 1994; 120: 2175-2186. 36. Grow, M.W. and Krieg, P. A. Tinman function is essential for vertebrate heart development: Elimination of cardiac differentiation by dominant inhibitory mutant of the tinman-related genes, XNkx2-3 and Xnkx25. Dev. Biol. 1998; 204: 187-196

7

and segmental borders. Development 1988: 104: 147160. 65. Martin, V.J., Littlefield, C.L., Archer, W.E. and Bode, H.R. Embryogenesis in hydra. Biol. Bull. 1997; 192: 345-363. 66. Zhang, J., King, M.L. Xenopus VegT RNA is localized to the vegetal cortex during oogenesis and encodes a novel T-box transcription factor involved in mesodermal patterning. Development 1996; 122: 4119-4129 67. Clements, D., Friday, R.V. and Woodland, H.R. (1999). Mode of action of VegT in mesoderm and endoderm formation. Development 1999; 126: 4903-4911. 68. Sun, B.I., Bush, S.M., Collins-Racie, L.A., LaVallie, E.R., DiBlasio-Smith, E.A., Wolfman, N.M., McCoy, J.M. and Sive, H.L. derriere: a TGF-beta family member required for posterior development in Xenopus. Development 1999; 126, 1467-1482. 69. Gierer, A., Berking, S., Bode, H., David, C.N., Flick, K., Hansmann, G., Schaller, H. and Trenkner, E. Regeneration of hydra from reaggregated cells. Nature New Biology 1972; 239, 98-101. 70. Meinhardt, H. A model for pattern-formation of hypostome, tentacles, and foot in hydra: how to form structures close to each other, how to form them at a distance. Dev. Biol. 1993; 157: 321-333. 71. Bally-Cuif, L. and Boncinelli, E. Transciption factors and head formation in vertebrates. BioEssay 1997; 19: 127-135. 72. Jones, C.M., Broadbent, J., Thomas, P.Q., Smith, J.C. and Beddington, R.S.P. An anterior signalling centre in Xenopus revealed by the homeobox gene XHex. Cur. Biol. 1999; 9: 946-954. 73. Niehrs, C. Head in the WNT - the molecular nature of Spemann’s head organizer. Trends Genetics 1999; 15: 314-319

53. Wasserman, J.D. and Freeman, M. An autoregulatory cascade of egf receptor signaling patterns the Drosophila egg. Cell 1998; 95: 355-364. 54. Nilson, L.A. and Sch¨ upbach, T. Localized requirements for windbeutel and pipe reveal a dorsoventral prepattern within the follicular epithelium of the Drosophila ovary. Cell 1998; 93, 253-262. 55. Nilson, L.A. and Sch¨ upbach, T.Egf receptor signaling in Drosophila oogenesis. Curr. Top. Dev. Biol 1999; 44, 203-243. 56. Sen, J., Goltz, J.S., Stevens, L. and Stein, D. Spatially restricted expression of pipe in the Drosophila egg chamber defines embryonic dorsal-ventral polarity. Cell 1998; 95: 471-481. 57. Chen, G., Handel, K. and Roth, S. The maternal nfkappa b/dorsal gradient of Tribolium castaneum: dynamics of early dorsoventral patterning in a short-germ beetle. Development 2000; 127,5145-5156. 58. Arendt, D. and N¨ ubler-Jung, K. Inversion of dorsoventral axis. Nature 1994; 371: 26-26. 59. De Robertis, E.M. and Sasai, Y. A common plan for dorsoventral patterning in bilateria. Nature 1996; 380: 37-40. 60. Lacalli, T. Dorsoventral axis inversion - a phylogenetic perspective. Bioessays 1996; 18, 251-254. 61. Singer, J.B., Harbecke, R., Kusch, T., Reuter, R. and Lengyel, J.A. Drosophila brachyenteron regulates gene activity and morphogenesis in the gut. Development 1996; 122: 3707-3718. 62. McGhee, J.D. Homologous tails? or tales of homology?. Bioessays 2000; 22: 781-785. 63. Weisblat, D.A., Price, D.J. and Wedeen, C.J. Segmentation in leech development. Development Supplement 1988; 104: 161-168. 64. Dohle, W. and Scholz, G. Clonal analysis of the crustacean segment: the discordance between genealogical

8